keraflow.layers.convolution.Convolution2D Class Reference

Convolution layer for convolving (input_depth, input_row, input_col) inputs. More...

Inheritance diagram for keraflow.layers.convolution.Convolution2D:

Public Member Functions
def	__init__
Public Member Functions inherited from keraflow.layers.convolution.ConvolutionBase
def	__init__

Detailed Description

Convolution layer for convolving (input_depth, input_row, input_col) inputs.

• input_shape: 4D, (nb_samples, input_depth, input_row, input_col)
• output_shape: 4D, (nb_samples, nb_kernel, output_row, output_col)
• parameters:
  • W: (input_depth, nb_kernel, input_row, input_col)
  • b: (nb_kernel,)

Note

output_row and output_col are determined pad and strides. For details, please see ConvolutionBase.

Constructor & Destructor Documentation

def keraflow.layers.convolution.Convolution2D.__init__	(		self,
			nb_kernel,
			kernel_row,
			kernel_col,
			strides = (1, 1,
			pad = 'valid',
			bias = True,
			init = 'glorot_uniform',
			activation = 'linear',
			kwargs
	)

Parameters

nb_kernel	int. Number of convolution kernels to use.
kernel_row	int. The height of the each kernel.
kernel_col	int. The width of the each kernel.
strides	2D tuple of int. Steps for vertically/horizontally sliding each kernel for convolution.
pad	str, 'valid' of 'same'. See ConvolutionBase.
bias	boolean. Whether to include a bias (i.e. make the layer affine rather than linear).
init	str/function. Function to initialize trainable parameters. See Initializations.
activation	str/function. Activation function applied on the output. See Activations.
kwargs	see Layer.__init__.